Editor Engagement/2013 strategy planning (Wikimedia Plan)

From mediawiki.org

This is a summary of Editor Engagement initiatives identified in the slides of Wikimedia Foundation's 2012-13 Annual Plan. These excerpts are included here as a reference for our editor engagement team's quarterly plan. See also: Engineering Goals 2012-13 and Editor Engagement Features (E2) 2013 strategy planning page.

Recapping 2011-12 Activities[edit]

Top priority: Editor Engagement

Objective: Increase participation on Wikimedia projects through small, iterative improvements and experiments which increase retention or inflow of new participants.

Original goal: Increase the number of active editors from just under 90K in March 2011, to 95K in June 2012.

Progress: Active editors continued to decline in 2011-12, reaching 85K in March 2012. Through the year, we divided our work into larger, foundational projects vs. small interventions and experiments. Larger projects launched in 2012 include:

  •  The MoodBar / Feedback Dashboard, a tool for new editors to report problems and frustrations, enabling experienced editors to respond;
  •  The Article Feedback Tool V5, which enables readers to submit free-text comments to improve articles and enables experienced editors to manage and resolve those comments;
  •  The New Pages Feed tool, a tool designed to help experienced editors manage the inflow of new page creations, while reducing friction and increasing gratitude expressed towards new users;
  •  The Teahouse (pilot project), a forum providing help and assistance to new editors

In addition, we completed various research projects and smaller interventions, e.g.:

  •  We completed the “Summer of Research” launched in the previous fiscal year, a comprehensive research program into editor dynamics;
  •  We completed a comprehensive study of the effects of different types of warning and welcoming messages on new editor retention;
  •  We ran first experiments on using e-mail to call back lapsed contributors.

In April 2012, we consolidated what was previously a departmentally split program of developers, researchers and community experts, into two teams: the Editor Engagement Features team (“E2”) and the Editor Engagement Experiments team (“E3”), both organized under the umbrella of the engineering and product department. We also began staffing up the E3 team with developers and design support.

Metrics and learnings associated with each initiative above are beyond the scope of this document. Ultimately the success of this program is defined by our ability to grow the number of active editors, and we have not yet done so. We believe that the consolidated and expanded teams will be able to achieve measurable impact on the active editor numbers in 2012-13, but we've scaled back our targets in light of the difficulty of the task.


2012-13 Risks[edit]

Editor decline continues to be an intractable problem.

Our projects' success depends on a thriving, diverse, healthy editing community, and yet editor numbers are declining. We believe this is due to:

  • insufficient site usability and discoverability of fun stuff to do
  •  increase in editorial warnings and reversions of new editors' work
  • declining support, hospitality and appreciation offered to newcomers by the community
  • growth in the editorial learning curve for newcomers due to increasing policy complexity. 

We believe sustained growth in editors will require a mix of small and large initiatives designed to address all these issues.

In 2011-12, the Wikimedia Foundation undertook a first set of initiatives targeting a reversal of the decline. Some, such as the -1 to 100 retention projects and editor recruitment initiatives in India and Brazil, were designed to have near-immediate pay-off, and others, such as the Visual Editor project, were not anticipated to yield results until post 2011-12. Our projections at the end of 2011-12 indicated that without intervention the number of active editors would have declined by March 2012 to 83.14K. Our target, based on the work we had planned, was 94.54K, and actuals for the month are 85.09K. Based on those numbers, it looks like editor decline has slowed, but has not yet stopped or reversed.

In 2012-13 therefore, we plan to redouble our efforts.

In 2012-13, we will continue the Visual Editor project, and expand our recruitment work in India, as well as investing further in in Brazil and MENA.

We will also launch projects designed to make contributing and curating multimedia easier, because multimedia is where early usability efforts (UploadWizard), especially alongside programs like Wiki Loves Monuments, have paid off. (Commons is one of the few areas where active editors are growing -- 25% year over year, with a spike to 9.37K from 6.97K in September 2011 due to the WLM competition.)

We will spend significant effort rearchitecting MediaWiki to support a responsive, engaging user experience, including developing a notification/feed system pushing updates to the user, and forming the foundation for a new messaging system. This, because we know messaging (talk pages) represents the other major barrier for new users besides wiki markup, and because virtually all other planned engagement initiatives will require better technological support for engaging directly with users.

Finally, acknowledging the complexity of the problem, and to ensure that we’re setting the right priorities with limited resources, we are increasing our investment in small research projects. For example in 2011-12 we ran an experiment e-mailing “lapsed” editors, to see whether such messages can bring them back. We will run experiments like this every week, and what we learn in them will inform other teams' editor retention work.


2012-13 Plan[edit]

Overview[edit]

The 2012-13 plan continues our focus on solving editor decline by doing work aimed at recruiting new editors and better retaining existing editors. 

We will do this primarily by adding engineering resources for specific initiatives, with some resources also being added for outreach work.

Key Activities[edit]

Editor Engagement Features[edit]

  • Develop a more responsive and engaging user experience. 
  • Implement shallow notifications system for user-relevant events by end of Q2. 
  • Launch new user-to-user messaging and scalable notifications system by end of Q4.

Editor Engagement Experiments[edit]

  • Support editor engagement work with rapid experimentation, and directly prioritize the most successful experiments. 
  • Conduct a minimum of 15 product and community experiments designed to directly increase new editor engagement and retention. 

Targets[edit]

Stabilize number of active editors (all projects except Commons) to 86,000 by July 1, 2013 from 85,000 in March 2012. We’re currently seeing a 1.5% y-o-y decline in active editors for all projects combined, and we believe reversing the decline will take sustained effort on multiple fronts. In 2012-13, we hope that the following initiatives will contribute materially to growing the community of (text) contributors:

  • editor engagement features
  • editor engagement experiments
  • visual editor
  • internationalization
  • site performance improvements.


Increase the number of contributors who make at least one upload to Wikimedia Commons from 18.6K in March 2012 to 25K in June 2013, including 1K mobile uploaders per month (from 0). There is a huge opportunity to increase the number of people donating images to the projects. The number of contributors to Commons grew by about 25% from March 2011 to March 2012, compared with ~12% in the prior year. We attribute this to two factors: The upload usability improvements made by the WMF (rolled out in May 2011), and the very successful “Wiki Loves Monuments” competition run by a number of chapters in fall 2011. In 2012-13, the WMF will for the first time provide a mobile app to support Wiki Loves Monuments, and will invest generally in mobile uploading and quality control. We will also continue to make usability and integration improvements to drive overall growth.

References[edit]